Question

older
Potential energy calculations with...

Paolo Daniele

7 Jan 2002 7 Jan '02

8:49 a.m.

Hi, I have a problem. Using Modeller program to make 100 models of my protein following the loop routine, the program was stopped because of an electric interruption, so I have now just 56 of 100 models! Is there a way to restart the program without loosing the 56 models yet obtained? Thanks in advance for your help. Best regards, Paolo Daniele

______________________________________________________________________

Iscriviti al gruppo ufficiale di Tomb Raider: http://it.groups.yahoo.com/group/gruppoufficiale_tombraider/

Show replies by date

Andrej Sali

7 Jan 7 Jan

9:15 a.m.

yes:

SET RAND_SEED = -48583 # some number < -1 and different from the default value SET STARTING_MODEL = 57 # change from 1

before you call 'model'

(similarly for the loop models, if that is what you are calculating instead of whole protein models)

best, andrej

-- Andrej Sali, Associate Professor The Rockefeller University, 1230 York Avenue, New York, NY 10021-6399 tel +1 212 327 7550; lab +1 212 327 7206 ; fax +1 212 327 7540 e-mail sali@rockefeller.edu; http://salilab.org

-----Original Message----- From: owner-modeller_usage@salilab.org [mailto:owner-modeller_usage@salilab.org]On Behalf Of Paolo Daniele Sent: Monday, January 07, 2002 11:50 AM To: modeller_usage@salilab.org Subject: Question

______________________________________________________________________

Iscriviti al gruppo ufficiale di Tomb Raider: http://it.groups.yahoo.com/group/gruppoufficiale_tombraider/

Björn Wallner

11 Jan 11 Jan

7:56 a.m.

New subject: Segmentation fault with modeller 6

Hi,

I've been using modeller 4 for about a year and just recently installed modeller 6. The problem I have is that I get segmentation fault when I try to run one of the example files for instance 'model-fast.top' in the exemples/tutorial-model directory.

I'll get the error when the program writes: "Restraint type to be calculated: distance"

An strace of the program before it crashes look like this:

write(1, "Dynamically allocated memory at "..., 94Dynamically allocated memory at amprmcns [B,kB,MB]: 6625069 6469.794 6.318 ) = 94 write(1, "make_re_417_> Restraint type to "..., 57make_re_417_> Restraint type to be calculated: distance ) = 57 stat64("5fd1", 0xbfff4cb8) = -1 ENOENT (No such file or directory) stat64("5fd1.Z", 0xbfff4cb8) = -1 ENOENT (No such file or directory) stat64("pdb5fd1", 0xbfff4cb8) = -1 ENOENT (No such file or directory) stat64("pdb5fd1.Z", 0xbfff4cb8) = -1 ENOENT (No such file or directory) stat64("./5fd1", 0xbfff4cb8) = -1 ENOENT (No such file or directory) stat64("./5fd1.Z", 0xbfff4cb8) = -1 ENOENT (No such file or directory) stat64("./pdb5fd1", 0xbfff4cb8) = -1 ENOENT (No such file or directory) stat64("./pdb5fd1.Z", 0xbfff4cb8) = -1 ENOENT (No such file or directory) stat64("./5fd1.atm", {st_mode=S_IFREG|0644, st_size=84641, ...}) = 0 brk(0x9571000) = 0x9571000 --- SIGSEGV (Segmentation fault) --- +++ killed by SIGSEGV +++

Any suggestions?

Sincerely Björn

-- Björn Wallner Stockholm Bioinformatics Center PhD Student SCFAB, Stockholm University bjorn@sbc.su.se S-106 91 Stockholm Tel:+46-8-553 785 66 Fax:+46-8-5537 8214 Cell:070-521 56 65

Christopher A. Wells

10:49 a.m.

New subject: chain breaks

I have been successfully using Mod4.0 to model multiple subunits together utilizing chain breaks in the alignment.

However this is causing problems in Mod6.0. I found in the manual for 6.0:

"For the time being, this and other alignment commands remove chain break information from the CALN array..." p.59

So this brings up several questions:

1) Is this the reason why I cannot create models with chain breaks in 6.0? 2) Is there an alternative solution besides sticking with Mod4.0 for the time being? 3) Is there an estimate on when chain breaks will be incorporated into Mod6.0?

Thanks in advance,

Christopher Wells MD/PhD Student Department of Pharmacology Medical University of South Carolina

Andras Fiser

16 Jan 16 Jan

2:22 p.m.

New subject: SUPPORTRe: chain breaks

HEllo Christopher,

Yes, it looks true.

In the older version (pre-release of mod6 and older) chainbreaks are OK in the alignment, in the current version (mod6v0) they cause problems. Only those chainbreak characters, that are within a sequence, not the ones, that separate heteroatoms.

On the other hand, you do not need to include them in the sequence, Modeller will automatically assume a chain break, if there is a longer than usual bond distance.

I hope, removing chain breaks from your seq will make your top file work, if not, let us know.

best wishes,

Andras

"Christopher A. Wells" wrote: > > I have been successfully using Mod4.0 to model multiple subunits > together utilizing chain breaks in the alignment. > > However this is causing problems in Mod6.0. I found in the manual for > 6.0: > > "For the time being, this and other alignment commands remove chain > break information from the CALN array..." p.59 > > So this brings up several questions: > > 1) Is this the reason why I cannot create models with chain breaks in > 6.0? > 2) Is there an alternative solution besides sticking with Mod4.0 for the > time being? > 3) Is there an estimate on when chain breaks will be incorporated into > Mod6.0? > > Thanks in advance, > > Christopher Wells > MD/PhD Student > Department of Pharmacology > Medical University of South Carolina

-- , Andras Fiser, PhD # phone: (212) 327 7216 The Rockefeller University # fax: (212) 327 7540 Box 270, 1230 York Avenue # e-mail:fisera@rockefeller.edu New York, NY 10021-6399, USA # http://salilab.org/~andras

8345

Age (days ago)

8354

Last active (days ago)

List overview

Download

4 comments

5 participants

tags (0)

participants (5)

Andras Fiser
Andrej Sali
Björn Wallner
Christopher A. Wells
Paolo Daniele